A method for predicting ornamental period of Pingyin rose
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JINAN METEOROLOGICAL BUREAU
- Filing Date
- 2026-03-03
- Publication Date
- 2026-06-19
Smart Images
Figure CN122242839A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of viewing period prediction, and in particular to a method for predicting the viewing period of Pingyin roses. Background Technology
[0002] Pingyin rose (Rosa rugosa var. plena) is a perennial deciduous shrub belonging to the genus Rosa in the family Rosaceae. It is one of the main cultivated rose varieties in China, with a cultivation history of over 1300 years. Pingyin roses are known for their rich fragrance, vibrant colors, large, thick petals, and excellent quality. They are important raw materials for traditional Chinese medicine, the food industry, and the fragrance industry, and are also high-quality ornamental materials for beautifying and greening the environment, earning them the title of "representative of traditional Chinese roses." The flowering time of roses is a key factor determining the harvesting and processing period and quality, and is also an important guide for leisure tourism in rose fields. Therefore, establishing a rose flowering period prediction model using years of data on the start of peak bloom and meteorological data, and providing meteorological services for the flowering period to predict the best viewing time, can provide a theoretical basis for tourists to rationally plan their viewing time and for the government's cultural tourism promotion and management work, fully realizing the potential economic, social, and ecological benefits of the rose flowering period.
[0003] Currently, there are reports from various countries on the prediction of flowering periods for various plants, most of which are based on correlation analysis between flowering period and meteorological factors. Chen et al. analyzed the changing characteristics of cherry blossom flowering period under the background of climate change and its relationship with average winter temperature; Han et al., through analysis of the correlation between flowering period and meteorological factors before flowering, found that March temperature is the main factor affecting the timing of the initial flowering of *Pyrus pyrifolia*; Bai et al. and Tan et al. also selected key meteorological factors affecting flowering period through correlation analysis and established multiple linear regression models to predict the initial flowering period of apples and the length of cherry blossom flowering, achieving certain results; Chen et al. established a stochastic dynamic model of peony flowering period and found that the closer the prediction date is to the initial flowering period, the more accurate the prediction.
[0004] Research on rose flowering period prediction has only been based on linear regression, without considering the nonlinear relationships between influencing factors. Currently, with the continuous development of computer technology, machine learning algorithms such as neural networks, support vector machines, random forests, and gradient boosting machines have been well applied in predicting the phenological periods and yields of plants such as pear blossoms and wheat. Among them, BP neural networks have strong nonlinear mapping capabilities. Summary of the Invention
[0005] The purpose of this application is to provide a method for predicting the viewing period of Pingyin roses, so as to improve the prediction accuracy of the beginning of the peak bloom of roses, and thus improve the prediction accuracy of the viewing period of roses.
[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for predicting the viewing period of Pingyin roses, including: Obtain daily meteorological data for the target area and target time period; the daily meteorological data includes average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine duration; Based on the daily meteorological data of the target region and the target time period, Pearson correlation analysis was used to identify meteorological factors that are significantly correlated with the peak blooming period of the rose to be predicted. Based on the meteorological factors that are highly significantly correlated with the peak blooming period of the rose to be predicted, the day sequence of the peak blooming period of the rose to be predicted is determined using a rose blooming period prediction model; wherein, the rose blooming period prediction model is obtained by training a BP neural network using a training dataset; The ornamental level of the rose to be tested is determined based on the date of the start of full bloom. The ornamental level includes a first-level ornamental period, a second-level ornamental period, and a third-level ornamental period. The first-level ornamental period is defined as the period from when more than 20% of the plants have fully opened flowers or when 50% of the plants have withered petals, based on the date of the start of full bloom. The second-level ornamental period is defined as the period from when more than 30% of the plants have fully opened flowers to when 30%-50% of the plants have withered petals, based on the date of the start of full bloom. The third-level ornamental period is defined as the period from when 5%-10% of the plants have fully opened flowers or when 50%-80% of the plants have withered petals, based on the date of the start of full bloom.
[0007] In one implementation, a highly significant correlation is defined as a Pearson correlation coefficient between -0.488 and -0.829.
[0008] In one embodiment, training the BP neural network using a training dataset specifically includes: Construct a training dataset and a BP neural network; the training dataset includes daily meteorological data for historical time periods of the target area and the true date sequence of the historical peak blooming period of roses; Based on daily meteorological data of the target area over a historical period, Pearson correlation analysis was used to identify meteorological factors that were significantly correlated with the historical peak blooming period of roses. Using the meteorological factors that are highly correlated with the historical peak blooming period of roses as input and the actual day number of the historical peak blooming period of roses as output, the BP neural network is trained using the LM optimization algorithm to obtain a rose blooming period prediction model.
[0009] In one embodiment, the BP neural network includes an input layer, a hidden layer, and an output layer; The number of nodes in the input layer is equal to the number of meteorological factors; The output layer has 1 node. The number of nodes in the hidden layer is determined based on the number of nodes in the input layer and the number of nodes in the output layer.
[0010] In one embodiment, determining the number of nodes in the hidden layer based on the number of nodes in the input layer and the number of nodes in the output layer specifically includes: Using formula Determine the range of the number of nodes in the hidden layer; where q is the number of nodes in the hidden layer; k is the number of nodes in the input layer; m is the number of nodes in the output layer; and a is a constant between [1, 10]. Based on the range of node numbers, a comparative experiment was conducted, and the number of nodes corresponding to the minimum training error was selected as the number of nodes in the hidden layer.
[0011] In one embodiment, during the training process, the learning rate is set to 0.01, the maximum number of training iterations is set to 1000, and the minimum error of the training target is set to 0.001.
[0012] Secondly, this application provides a system for predicting the viewing period of Pingyin roses. This system is used to implement the aforementioned method for predicting the viewing period of Pingyin roses. The system includes: The data acquisition module is used to acquire daily meteorological data for a target area and a target time period; the daily meteorological data includes average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine duration; The data filtering module is used to determine meteorological factors that are significantly correlated with the peak blooming period of the rose to be predicted, based on the daily meteorological data of the target region and the target time period and using Pearson correlation analysis. The flowering period prediction module is used to determine the date sequence of the peak flowering period of the rose to be predicted based on the meteorological factors that are highly significantly related to the peak flowering period of the rose to be predicted, using a rose flowering period prediction model; wherein, the rose flowering period prediction model is obtained by training a BP neural network using a training dataset; The viewing period grading module is used to determine the viewing level of the rose under test based on the date of the start of full bloom. The viewing level includes a first-level viewing period, a second-level viewing period, and a third-level viewing period. The first-level viewing period is the period from when more than 20% of the plants have fully opened flowers or 50% of the plants have withered petals, based on the date of the start of full bloom. The second-level viewing period is the period from when more than 30% of the plants have fully opened flowers to when 30%-50% of the plants have withered petals, based on the date of the start of full bloom. The third-level viewing period is the period from when 5%-10% of the plants have fully opened flowers or 50%-80% of the plants have withered petals, based on the date of the start of full bloom.
[0013] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for predicting the viewing period of Pingyin roses.
[0014] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for predicting the viewing period of Pingyin roses.
[0015] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for predicting the viewing period of Pingyin roses.
[0016] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method for predicting the viewing period of Pingyin roses. It involves acquiring daily meteorological data for a target area and a target time period. The daily meteorological data includes average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine duration. Based on the daily meteorological data for the target area and the target time period, Pearson correlation analysis is used to determine meteorological factors that are significantly correlated with the peak blooming period of the roses to be predicted. Based on these meteorological factors, a rose blooming period prediction model is used to determine the date sequence of the peak blooming period of the roses to be predicted. The rose blooming period prediction model is obtained by training a BP neural network using a training dataset. In this application, the meteorological factors selected through correlation analysis are used as the input layer (independent variables), and the date sequence of the peak blooming period is used as the output layer (dependent variable). A BP neural network is used to construct a meteorological prediction model for the peak blooming period of roses (rose blooming period prediction model) to improve the prediction accuracy of the peak blooming period, thereby improving the prediction accuracy of the rose viewing period. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A flowchart illustrating a method for predicting the viewing period of Pingyin roses, provided as an embodiment of this application; Figure 2 This is a diagram illustrating the viewing period levels; Figure 3 A graph showing the trend of the peak blooming period of Pingyin roses from 1994 to 2024; Figure 4 A graph showing the back-substitution test results of the BP neural network model at the beginning of rose bloom; Figure 5 A graph showing the back-substitution test results of a stepwise multiple linear regression model for the initial blooming period of roses; Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0020] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0021] In one exemplary embodiment, such as Figure 1 As shown, a method for predicting the viewing period of Pingyin roses is provided, including the following steps: S1: Obtain daily meteorological data for the target area and target time period; the daily meteorological data includes average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine duration.
[0022] S2: Based on the daily meteorological data for the target region and the target time period, Pearson correlation analysis is used to determine the meteorological factors that are highly significantly correlated with the onset of full bloom of the roses to be predicted. Highly significant correlation is defined as a Pearson correlation coefficient between -0.488 and -0.829.
[0023] S3: Based on the meteorological factors that are highly significantly related to the peak blooming period of the rose to be predicted, the date sequence of the peak blooming period of the rose to be predicted is determined using a rose blooming period prediction model; wherein, the rose blooming period prediction model is obtained by training a BP neural network using a training dataset.
[0024] In this embodiment, the BP neural network includes an input layer, a hidden layer, and an output layer; the number of nodes in the input layer is equal to the number of meteorological factors; the number of nodes in the output layer is 1; the number of nodes in the hidden layer is determined based on the number of nodes in the input layer and the number of nodes in the output layer.
[0025] In this embodiment, determining the number of nodes in the hidden layer based on the number of nodes in the input layer and the number of nodes in the output layer specifically includes: Using formula Determine the range of the number of nodes in the hidden layer; where q is the number of nodes in the hidden layer; k is the number of nodes in the input layer; m is the number of nodes in the output layer; and a is a constant between [1, 10].
[0026] Based on the range of node numbers, a comparative experiment was conducted, and the number of nodes corresponding to the minimum training error was selected as the number of nodes in the hidden layer.
[0027] S4: Determine the viewing level of the rose to be tested based on the date of the start of full bloom; the viewing level includes a first-level viewing period, a second-level viewing period, and a third-level viewing period; the first-level viewing period is the period from when more than 20% of the plants have fully opened flowers or 50% of the plants have withered petals, based on the date of the start of full bloom; the second-level viewing period is the period from when more than 30% of the plants have fully opened flowers to when 30%-50% of the plants have withered petals, based on the date of the start of full bloom; the third-level viewing period is the period from when 5%-10% of the plants have fully opened flowers or 50%-80% of the plants have withered petals, based on the date of the start of full bloom.
[0028] A diagram illustrating the viewing period levels is shown below. Figure 2 As shown in Table 1, the viewing period is a specific phenological phenomenon that has ornamental value. Based on the characteristics of rose blooming, a grading standard for the viewing period of ornamental roses has been established, providing scientific and technological support for refined meteorological services such as government decision-making, cultural and tourism activities, and public flower viewing experiences.
[0029] Table 1. Grading Standards for the Ornamental Rose Viewing Period
[0030] In this embodiment, training the BP neural network using the training dataset specifically includes: Step 1: Construct a training dataset and a BP neural network; the training dataset includes daily meteorological data for historical time periods of the target region and the true date sequence of the historical peak blooming period of roses.
[0031] First, the study area was selected. In this embodiment, an area located between 36°1′—36°23′N and 116°12′—116°37′E, with an altitude of 100–250m, was chosen as the study area. The soil in the study area is mainly brown soil, belonging to the warm temperate continental semi-humid monsoon climate zone, with distinct seasons, abundant sunshine, and concentrated rainfall. This unique geographical location, suitable soil, and climate have created the beautiful and fragrant Pingyin rose. As of 2019, the total rose planting area in Pingyin County was 1762 hm². 2 The total output of dried flowers reached 1788 tons.
[0032] The training dataset was obtained from the observation data of the beginning of the peak bloom of roses from 1994 to 2024, which came from rose planting bases and meteorological bureaus in the study area; and from early January to mid-April from 1994 to 2024, the daily meteorological data (average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine hours) came from national reference stations located in the study area.
[0033] BP neural network construction method: The BP (Back Propagation) neural network is a multi-layer feedforward network based on the backpropagation algorithm. Its topology consists of an input layer, hidden layers, and an output layer, with each layer containing several nodes. The selection of the number of nodes in the hidden layer is crucial. In this embodiment, the range of values for the number of nodes in the hidden layer is calculated using the empirical formula (1). A BP neural network prediction model is established using meteorological factors as the input layer and the date of the beginning of full bloom as the output layer.
[0034] (1) In the formula, q is the number of hidden layer nodes; k is the number of input layer nodes; m is the number of output layer nodes; and a is a constant between [1, 10].
[0035] Step 2: Based on the daily meteorological data of the target area over a historical period, use Pearson correlation analysis to determine the meteorological factors that are highly significantly correlated with the historical peak blooming period of roses.
[0036] In this embodiment, from 1994 to 2024, the average date number of the start of full bloom for Pingyin roses in the study area was 121, corresponding to May 1st (common year) or April 30th (leap year). The earliest start of full bloom was April 26th (2002, 2004, and 2016), and the latest was May 7th, 1996 (leap year) and May 8th, 2010 (common year), with a difference of 11 days. Statistically, the start of full bloom occurred in early May in 16 years, accounting for 51.6% of the total sample; and in late April in 15 years, accounting for 48.4%. Based on the least squares method, the changing trend of the date number of the start of full bloom for Pingyin roses from 1994 to 2024 was analyzed, as follows... Figure 3 As shown, the results indicate that the date of the start of full bloom shows a fluctuating decreasing trend, meaning that the start of full bloom is trending earlier, averaging 0.4 days earlier every 10 years.
[0037] Given the widespread practice of artificial irrigation before flowering and the poor controllability of precipitation and soil moisture, their use as predictive factors carries significant uncertainty. Therefore, a correlation analysis was conducted on 126 meteorological factors, including monthly and ten-day average temperatures, average maximum temperatures, average minimum temperatures, average 5cm soil temperature, sunshine duration, effective accumulated temperature ≥0℃, ≥3℃, ≥5℃, and ≥10℃, from January to mid-April between 1994 and 2020, and the date of full bloom commencement. As shown in Table 2, the results indicate that 33 meteorological factors showed significant correlations with the date of full bloom commencement (P<0.05), of which 16 showed highly significant correlations (P<0.01), with correlation coefficients ranging from -0.488 to -0.829. As shown in Table 2, the timing of the peak bloom of Pingyin roses is mainly influenced by meteorological factors in early to mid-April. Temperature has a more significant impact on the peak bloom time than sunshine duration. Among these factors, the average maximum temperature has a more significant impact than the minimum temperature, and the average 5 cm soil temperature in mid-April has the most significant impact on the peak bloom time. Among the effective accumulated temperature factors, the correlation coefficients between monthly and ten-day effective accumulated temperatures ≥0℃ and ≥3℃ and the peak bloom time are similar or consistent, while the correlation coefficients between effective accumulated temperatures ≥5℃ and ≥10℃ and the peak bloom time are both higher. Sixteen meteorological factors whose correlation with the model passed the highly significant test were selected as the meteorological factors for establishing the prediction model of the beginning of the flowering period. These factors are: average temperature in early April (T1), average temperature in mid-April (T2), average maximum temperature in mid-April (X1), average minimum temperature in mid-April (N1), average 5cm soil temperature in early April (G1), average 5cm soil temperature in mid-April (G2), sunshine hours in mid-April (S1), effective accumulated temperature ≥0℃ in early April (E1), effective accumulated temperature ≥0℃ in mid-April (E2), effective accumulated temperature ≥3℃ in early April (E3), effective accumulated temperature ≥3℃ in mid-April (E4), effective accumulated temperature ≥5℃ in March (E5), effective accumulated temperature ≥5℃ in early April (E6), effective accumulated temperature ≥5℃ in mid-April (E7), effective accumulated temperature ≥10℃ in early April (E8), and effective accumulated temperature ≥10℃ in mid-April (E9).
[0038] Table 2. Correlation analysis of monthly and ten-day meteorological factors and the date sequence of the beginning of full bloom from mid-January to mid-April, 1994 to 2020.
[0039] in, , The '-' indicates that the significance test was passed at the 0.05 and 0.01 levels, respectively, and the '-' indicates that at least one variable is a constant and therefore cannot be calculated.
[0040] Finally, the BP neural network was constructed: the date sequence of the start of full bloom of roses from 1994 to 2020 was selected as the output layer of the BP neural network, i.e., 1 node; 16 meteorological factors that are highly significantly correlated with the date sequence of the start of full bloom were selected as the input layer of the BP neural network, i.e., 16 nodes; the selection range of the number of hidden layer nodes was calculated to be [5,14] by formula (1), as shown in Table 3. Through comparative experiments, it was found that when the number of hidden layer nodes was 10, the training error was the smallest (0.039) and the training result was the best; therefore, the topology of the BP neural network was determined to be 16-10-1 (input layer, hidden layer, output layer). During the training process, the hyperbolic tangent function (tansig) was used as the activation function for the hidden layer, the Levenberg-Marquardt optimization algorithm (trainlm) was used for the training algorithm, the learning rate was set to 0.01, the maximum number of training iterations was set to 1000, the minimum training error was set to 0.001, and the linear transfer function (purelin) was used for the output layer.
[0041] Figure 4 The simulated rounding result of the selected optimal model on the training set, such as Figure 4 As shown, the predicted start date of full bloom was exactly the same as the actual start date in 17 years, accounting for 63.0% of the total sample; the prediction error was ±1 day in 8 years, accounting for 29.6%; the prediction error was ±2 days in 2 years, accounting for 7.4%; and the average absolute error of the model prediction was 0.44 days.
[0042] Table 3 Training Error Table for Different Number of Hidden Layer Nodes
[0043] Step 3: Using the meteorological factors that are highly significantly correlated with the historical peak blooming period of roses as input and the actual day sequence number of the historical peak blooming period of roses as output, train the BP neural network using the LM optimization algorithm to obtain the rose blooming period prediction model.
[0044] In this embodiment, a stepwise multiple linear regression model was also established. Stepwise multiple linear regression (SMLR) is a method that predicts the dependent variable by analyzing the correlation between a single dependent variable and multiple independent variables and selecting the optimal combination of independent variables. Its principle is to introduce independent variables that significantly affect the dependent variable one by one, and then re-examine all independent variables in the equation, eliminating insignificant independent variables one by one, thereby selecting the most statistically significant independent variables and establishing the regression equation. In this embodiment, SPSS 25.0 was used, employing a stepwise method to select independent variables; the variance inflation factor (VIF) was used to test for multicollinearity in the regression equation; a VIF > 10 indicates the presence of multicollinearity; and an F-test was performed; if F > F<0.01, the regression model is considered effective.
[0045] Establishment of a stepwise multiple linear regression model: The date sequence of the start of full bloom of roses from 1994 to 2020 was selected as the dependent variable, and 11 meteorological factors that were significantly correlated with the date sequence of the start of full bloom were used as the initial set of independent variables. The stepwise multiple linear regression method was used for modeling. Three meteorological factors were selected as predictive factors through the stepwise method: mid-April 5cm ground temperature (G2), early April ≥10℃ effective accumulated temperature (E8), and mid-April sunshine hours (S1). A stepwise multiple linear regression model was established, as shown in Equation (2).
[0046] Y=142.053-0.752G2-0.077E8-0.064S1(2) In the formula, Y is the date number of the beginning of full bloom; G2 is the soil temperature at 5cm depth in mid-April; E8 is the effective accumulated temperature of ≥10℃ in early April; and S1 is the sunshine duration in mid-April.
[0047] Tables 4 and 5 show the test results for the model coefficients and the analysis of variance, respectively. As shown in Table 4, the t-test results indicate that all independent variables in the model reach a significance level, with a tolerance greater than 0.1 and a VIF value less than 10, indicating that there is no multicollinearity among the independent variables. The analysis of variance results are shown in Table 5. The F-value is 34.218, which is greater than the critical value F0.01(3,23)=4.765, and P<0.01, indicating that the model passes the F-test, the regression is significant, and it can be used for actual prediction.
[0048] Table 4. Model Coefficient Test Results
[0049] Table 5. Results of Model ANOVA Test
[0050] Model back-substitution test. Based on the established stepwise multiple linear regression prediction model, the date sequence of the start of full bloom for Pingyin roses from 1994 to 2020 was simulated, and the simulation results were rounded. The results are as follows: Figure 5 As shown, the predicted date of full bloom was consistent with the actual date of full bloom in 8 years, accounting for 29.6% of the total sample; the prediction error was ±1 day or ±2 days in 17 years, accounting for 63.0%; the prediction error was 3 days in only 2 years, accounting for 7.4%; the average absolute error of the model prediction was 1.04 days.
[0051] Model validation methods: Using the root mean square error (RMSE), relative error (RE), and coefficient of determination R0 2 Accuracy evaluation and error analysis are performed on the prediction model.
[0052] (3) (4) (5) In the formula, n represents the number of years in which Pingyin roses begin to bloom; y represents the date number of the actual blooming start date. To predict the date of the start of full bloom; This represents the average number of actual days at the start of full bloom. The smaller the values of RMSE and RE, the smaller the deviation between the predicted and actual values, and the higher the model accuracy. 2 It can measure how well a prediction model fits the actual values; the closer its value is to 1, the better the model performs.
[0053] Comparison of prediction accuracy between the two models: By calculating RMSE, RE, and R 2 The prediction accuracy of the BP neural network model and the stepwise multiple linear regression model were compared. Table 6 shows that the RMSE of the BP neural network prediction model on the training set was 0.75 days, and the RE was 0.62%, which are lower than the RMSE (1.31 days) and RE (1.08%) of the stepwise multiple linear regression model, indicating that the prediction model for the beginning of flowering based on the BP neural network has a smaller error. The RMSE of the BP neural network model on the training set was... 2 The R-value was 0.92, higher than that of the stepwise multiple linear regression model. 2 The value (0.77) indicates that the prediction model based on the BP neural network has a higher fit to the trend of rose bloom at the beginning of flowering. In summary, the BP neural network model outperforms the traditional stepwise multiple linear regression model in both prediction accuracy and fitting effect.
[0054] Table 6 Comparison of the accuracy of the two models
[0055] Comparison of the trial reporting results of the two models: The prediction performance of the BP neural network model and the stepwise multiple linear regression model was validated using data on the start of full bloom for Pingyin roses from 2021 to 2024, along with meteorological data. The results are shown in Table 7. From 2021 to 2024, the BP neural network model predicted the start of full bloom in 75.0% of the years (2021, 2023, and 2024), with only 2022 showing a one-day absolute error. The stepwise multiple linear regression model predicted the start of full bloom in only 25.0% of the years (2023), with an absolute error of one day in 2021 and 2022, and two days in 2024. Overall, both models can predict the start of full bloom for Pingyin roses relatively well, with small average absolute errors. The comparison shows that the BP neural network-based prediction model is three times more accurate than the stepwise multiple linear regression model, with a 75.0% reduction in average absolute error and a 50.0% reduction in maximum absolute error, demonstrating higher accuracy and better prediction performance.
[0056] Table 7. Validation results of the prediction performance of the two models.
[0057] Verification conclusion: (1) Preliminary screening of meteorological factors is the foundation for constructing a flowering period prediction model and an effective method to improve prediction accuracy. In this application, 16 meteorological factors that are highly significantly correlated with the date number of the start of full bloom were screened through Pearson correlation analysis to establish a meteorological prediction model. Among all meteorological factors, the average 5cm ground temperature and average maximum air temperature in mid-April have the highest negative correlation with the date number of the start of full bloom, indicating that the higher the temperature in the early stage, the earlier the start of full bloom. This also shows that the temperature factor in mid-April is an important factor affecting the timing of the start of full bloom of roses.
[0058] (2) The choice of modeling method is an important factor affecting the accuracy of rose blooming time prediction. Stepwise multiple linear regression is relatively simple and intuitive, and has stronger operability in practical applications. Compared with multiple linear regression, it improves prediction accuracy and reliability. However, it may ignore some independent variables that have practical significance but do not contribute much to the model. Different datasets may lead to different independent variables being selected by the model. Therefore, the modeling results of stepwise multiple linear regression may lack stability, and the accuracy will also be affected to some extent. Backpropagation (BP) neural networks can effectively capture the complex nonlinear relationship between input and output. After proper training, the neural network can also generalize well to new datasets. If the modeling data is sufficiently representative, the BP neural network can build a model with better prediction effect by virtue of its own optimization computing power. The prediction results of this application on the blooming time of Pingyin roses in Pingyin County show that the prediction accuracy of the BP neural network prediction model is better than that of the stepwise multiple linear regression model.
[0059] (3) Both models constructed in this application can predict the start of full bloom for Pingyin roses in late April (6 to 17 days earlier), but they are only applicable to the study area. The timing of the start of full bloom for roses is not only related to meteorological factors, but also to meteorological disasters before flowering, field management, fertilizer application, and other factors. Therefore, the prediction results of the two models still have certain errors, and in practical applications, they need to be combined with the specific conditions in the field. In the future, it is necessary to further collect rose phenological observation data, continuously study and improve meteorological prediction methods, and make the prediction models more practical and representative.
[0060] This application uses meteorological factors selected through correlation analysis as the input layer (independent variables) and the date sequence of the peak bloom as the output layer (dependent variable). A rose bloom prediction model is constructed using a BP neural network and compared with a stepwise multiple linear regression forecast model to improve the prediction accuracy of the rose bloom prediction model. This aims to provide technical support for the promotion of Pingyin roses and the harvesting of roses by flower farmers. In addition, accurately dividing the blooming period, making accurate forecasts of the blooming period, and recommending suitable blooming spots can create new hotspots for tourism consumption and promote the deep integration of the rose industry with culture and tourism.
[0061] Based on the same inventive concept, this application also provides a system for predicting the viewing period of Pingyin roses, which is used to implement the above-mentioned method for predicting the viewing period of Pingyin roses. The solution provided by this system is similar to the solution described in the above-described method. Therefore, the specific limitations in the embodiments of the Pingyin rose viewing period prediction system provided below can be found in the limitations of the Pingyin rose viewing period prediction method described above, and will not be repeated here.
[0062] In one exemplary embodiment, a system for predicting the viewing period of Pingyin roses is provided, comprising: The data acquisition module is used to acquire daily meteorological data for a target area and a target time period; the daily meteorological data includes average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine duration.
[0063] The data filtering module is used to determine meteorological factors that are significantly correlated with the peak blooming period of the rose to be predicted, based on the daily meteorological data of the target region and the target time period, using Pearson correlation analysis.
[0064] The flowering period prediction module is used to determine the date sequence of the peak flowering period of the rose to be predicted based on the meteorological factors that are highly significantly related to the peak flowering period of the rose to be predicted, using a rose flowering period prediction model; wherein, the rose flowering period prediction model is obtained by training a BP neural network using a training dataset.
[0065] The viewing period grading module is used to determine the viewing level of the rose under test based on the date of the start of full bloom. The viewing level includes a first-level viewing period, a second-level viewing period, and a third-level viewing period. The first-level viewing period is the period from when more than 20% of the plants have fully opened flowers or 50% of the plants have withered petals, based on the date of the start of full bloom. The second-level viewing period is the period from when more than 30% of the plants have fully opened flowers to when 30%-50% of the plants have withered petals, based on the date of the start of full bloom. The third-level viewing period is the period from when 5%-10% of the plants have fully opened flowers or 50%-80% of the plants have withered petals, based on the date of the start of full bloom.
[0066] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method for predicting the viewing period of Pingyin roses.
[0067] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the above-described method for predicting the viewing period of Pingyin roses.
[0068] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the above-described method for predicting the viewing period of Pingyin roses.
[0069] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for predicting the viewing period of Pingyin roses.
[0070] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0071] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.
[0072] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
[0073] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0074] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0075] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method for predicting the viewing period of Pingyin roses, characterized in that, include: Obtain daily meteorological data for the target area and target time period; the daily meteorological data includes average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine duration; Based on the daily meteorological data of the target region and the target time period, Pearson correlation analysis was used to identify meteorological factors that are significantly correlated with the peak blooming period of the rose to be predicted. Based on the meteorological factors that are highly significantly correlated with the peak blooming period of the rose to be predicted, the day sequence of the peak blooming period of the rose to be predicted is determined using a rose blooming period prediction model; wherein, the rose blooming period prediction model is obtained by training a BP neural network using a training dataset; The ornamental level of the rose to be tested is determined based on the date of the start of full bloom. The ornamental level includes a first-level ornamental period, a second-level ornamental period, and a third-level ornamental period. The first-level ornamental period is defined as the period from when more than 20% of the plants have fully opened flowers or when 50% of the plants have withered petals, based on the date of the start of full bloom. The second-level ornamental period is defined as the period from when more than 30% of the plants have fully opened flowers to when 30%-50% of the plants have withered petals, based on the date of the start of full bloom. The third-level ornamental period is defined as the period from when 5%-10% of the plants have fully opened flowers or when 50%-80% of the plants have withered petals, based on the date of the start of full bloom.
2. The method for predicting the viewing period of Pingyin roses according to claim 1, characterized in that, Highly significant correlation is defined as a Pearson correlation coefficient between -0.488 and -0.
829.
3. The method for predicting the viewing period of Pingyin roses according to claim 2, characterized in that, Training a backpropagation (BP) neural network using a training dataset specifically includes: Construct a training dataset and a BP neural network; the training dataset includes daily meteorological data for historical time periods of the target area and the true date sequence of the historical peak blooming period of roses; Based on daily meteorological data of the target area over a historical period, Pearson correlation analysis was used to identify meteorological factors that were significantly correlated with the historical peak blooming period of roses. Using the meteorological factors that are highly correlated with the historical peak blooming period of roses as input and the actual day number of the historical peak blooming period of roses as output, the BP neural network is trained using the LM optimization algorithm to obtain a rose blooming period prediction model.
4. The method for predicting the viewing period of Pingyin roses according to claim 3, characterized in that, The BP neural network includes an input layer, hidden layers, and an output layer; The number of nodes in the input layer is equal to the number of meteorological factors; The output layer has 1 node. The number of nodes in the hidden layer is determined based on the number of nodes in the input layer and the number of nodes in the output layer.
5. The method for predicting the viewing period of Pingyin roses according to claim 4, characterized in that, The number of nodes in the hidden layer is determined based on the number of nodes in the input layer and the number of nodes in the output layer, specifically including: Using formula Determine the range of the number of nodes in the hidden layer; where q is the number of nodes in the hidden layer; k is the number of nodes in the input layer; m is the number of nodes in the output layer; and a is a constant between [1, 10]. Based on the range of node numbers, a comparative experiment was conducted, and the number of nodes corresponding to the minimum training error was selected as the number of nodes in the hidden layer.
6. The method for predicting the viewing period of Pingyin roses according to claim 3, characterized in that, During training, the learning rate was set to 0.01, the maximum number of training iterations was set to 1000, and the minimum error of the training target was set to 0.
001.
7. A system for predicting the viewing period of Pingyin roses, characterized in that, The Pingyin rose viewing period prediction system is used to implement the Pingyin rose viewing period prediction method according to any one of claims 1-6, and the Pingyin rose viewing period prediction system includes: The data acquisition module is used to acquire daily meteorological data for a target area and a target time period; the daily meteorological data includes average temperature, maximum temperature, minimum temperature, 5cm ground temperature, and sunshine duration; The data filtering module is used to determine meteorological factors that are significantly correlated with the peak blooming period of the rose to be predicted, based on the daily meteorological data of the target region and the target time period and using Pearson correlation analysis. The flowering period prediction module is used to determine the date sequence of the peak flowering period of the rose to be predicted based on the meteorological factors that are highly significantly related to the peak flowering period of the rose to be predicted, using a rose flowering period prediction model; wherein, the rose flowering period prediction model is obtained by training a BP neural network using a training dataset; The viewing period grading module is used to determine the viewing level of the rose under test based on the date of the start of full bloom. The viewing level includes a first-level viewing period, a second-level viewing period, and a third-level viewing period. The first-level viewing period is the period from when more than 20% of the plants have fully opened flowers or 50% of the plants have withered petals, based on the date of the start of full bloom. The second-level viewing period is the period from when more than 30% of the plants have fully opened flowers to when 30%-50% of the plants have withered petals, based on the date of the start of full bloom. The third-level viewing period is the period from when 5%-10% of the plants have fully opened flowers or 50%-80% of the plants have withered petals, based on the date of the start of full bloom.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for predicting the viewing period of Pingyin roses according to any one of claims 1-6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for predicting the viewing period of Pingyin roses as described in any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for predicting the viewing period of Pingyin roses as described in any one of claims 1-6.